Relaxing Constraints due to Data and Control Dependences

スポンサーリンク

概要

• 論文の詳細を見る

• TLSP (Thread-Level Speculative Parallel pro-cessing) archtecture is a growing processor architecture. Parallelism of a program executed on this architecture is ruled by the combination of techniques which relax data dependences. In this paper, we evaluate the limits of parallelism of the TLSP architecture by using abstract machine models. We have three major results. First, if we use solely each technique which relaxes data dependences, "renaming" has a large effect on the TLSP architecture. Second, combinatiorial use of "memory disambiguation" and "renaming" leads to huge parallelism. Third, constant effects are obtained by concurrent use of "value prediction" and other techniques.

• 社団法人電子情報通信学会の論文
• 2003-05-01

著者

• Murakami Kazuaki

  Division Of Informatics Kyushu University

• METSUGI Katsuhiko

  Department of Informatics Kyushu University

関連論文

• Relaxing Constraints due to Data and Control Dependences
• Relaxing Constraints due to Data and Control Dependences

スポンサーリンク